Professional Documents
Culture Documents
Abstract
DNA methylation regulates gene expression and can be modified by different bioactive compounds in foods, such as
polyphenols. Cocoa is a rich source of polyphenols, but its role in DNA methylation is still unknown. The objective was to
assess the effect of cocoa consumption on DNA methylation and to determine whether the enzymes involved in the DNA
methylation process participate in the mechanisms by which cocoa exerts these effects in humans. The global DNA
methylation levels in the peripheral blood were evaluated in 214 volunteers who were pre-hypertensive, stage-1
hypertensive or hypercholesterolemic. The volunteers were divided into two groups: 110 subjects who consumed cocoa
(6 g/d) for two weeks and 104 control subjects. In addition, the peripheral blood mononuclear cells (PBMCs) from six
subjects were treated with a cocoa extract to analyze the mRNA levels of the DNA methyltransferases (DNMTs),
methylenetetrahydrofolate reductase (MTHFR), and methionine synthase reductase (MTRR) genes. Cocoa consumption
significantly reduced the DNA methylation levels (2.99160.366 vs. 3.90960.380, p,0.001). Additionally, we found an
association between the cocoa effects on DNA methylation and three polymorphisms located in the MTHFR, MTRR, and
DNMT3B genes. Furthermore, in PBMCs, the cocoa extract significantly lowered the mRNA levels of the DNMTs, MTHFR, and
MTRR. Our study demonstrates for the first time that the consumption of cocoa decreases the global DNA methylation of
peripheral leukocytes in humans with cardiovascular risk factors. In vitro experiments with PBMCs suggest that cocoa may
exert this effect partially via the down-regulation of DNMTs, MTHFR and MTRR, which are key genes involved in this
epigenetic process.
Citation: Crescenti A, Solà R, Valls RM, Caimari A, del Bas JM, et al. (2013) Cocoa Consumption Alters the Global DNA Methylation of Peripheral Leukocytes in
Humans with Cardiovascular Disease Risk Factors: A Randomized Controlled Trial. PLoS ONE 8(6): e65744. doi:10.1371/journal.pone.0065744
Editor: Rajasingh Johnson, University of Kansas Medical Center, United States of America
Received October 5, 2012; Accepted April 26, 2013; Published June 26, 2013
Copyright: ß 2013 Crescenti et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by ACC1Ó (TECCT11-1-0012) and the Ministerio de Economı́a y Competitividad (AGL2008-00387). The funders had no role in
study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: Anna Anguera is an employee of Rottapharm/Madaus, S.L., and Neus Anglés is an employee of La Morella Nuts, S.A.U. This does not alter
the authors9 adherence to all the PLOS ONE policies on sharing data and materials.
* E-mail: anna.crescenti@ctns.cat.
performed in cells have demonstrated that these bioactive food .35 Kg/m2, and a history of CVD. Participant eligibility or
components exhibit cancer inhibition activities by affecting exclusion was assessed by the attending physician and was based of
epigenetic signaling pathways, including DNA methylation [2]. review of clinical records, followed by screening visits [21,22].
For instance, it has been shown that epigallocatechin-3-gallate
(EGCG), the major polyphenol in green tea, and the soybean Study characteristics
isoflavone genistein can reverse DNA hypermethylation and For the cocoa group (45 men and 65 women), during a
increase the expression of various methylated-silenced tumor stabilization period of 2 weeks, all participants received the cocoa
suppressor (TS) genes, such as methylguanine methyltransferase, cream product (composition as described in Table 1) in order to
p16INK4a, human MutL Homolog 1 and retinoic acid receptor b, in assess the subject’s tolerance to the product (13 g/unit; 1 g cocoa/
different human cancer cells lines [10–12]. In addition to EGCG unit, 6 units/d; 6 g cocoa/d; 465 Kcal/d) included in an isocaloric
and genistein, other dietary polyphenolic compounds, such as diet with 13% of total energy as saturated fatty acids (treated
catechin, epicatechin, caffeic acid, chlorogenic acid, and curcu- group) [22]. For the control group (46 men and 58 women), in the
min, have been shown to have similar effects on DNA methylation run-in period of 2 weeks duration, the participants consumed an
[10,13,14]. Nevertheless, the evidence for a modulating effect of isocaloric diet in which the percentage of saturated fatty acids in
dietary polyphenols on DNA methylation in vivo is less convincing the diet was 13% of total energy [21].
[15–18]. Fasting blood samples from participants were obtained at the
Cocoa is a rich source of polyphenols and is a significant end of 2 weeks in each study for the isolation of peripheral
contributor to the total dietary intake of these compounds in leukocyte DNA.
humans [19,20]. However, the effect of cocoa on DNA All participants provided written informed consent prior to
methylation has not been addressed previously. Our hypothesis enrolment in the trial.
is that cocoa polyphenols reduces global DNA methylation and For the in vitro study, PBMCs were obtained from six adult
that this reduction is mediated via the regulation of key genes volunteers (3 men and 3 women, aged 30–34 years) with their oral
involved in this epigenetic process.
informed consent.
The aim of the present study was to assess the effect of cocoa on
DNA methylation and to determine whether the enzymes involved
in the DNA methylation process participate in the mechanisms by
Ethics statement
which cocoa exerts these effects in humans. To do so, we studied The study was approved by the Clinical Research Ethical
the effect of cocoa consumption on the global 5mC levels of Committee of the Hospital Universitary Sant Joan de Reus and at 3
peripheral leukocyte DNA in a population with cardiovascular risk Primary-Care Centers (Alcover, Vic, Centelles) where participants
factors. Additionally, to identify the possible mechanism by which were recruited and the human experimentation was conducted.
cocoa consumption influences DNA methylation, we investigated This study was conducted according to the guidelines established
whether genetic variants of the enzymes involved in the provision in the Declaration of Helsinki. The trial was registered with
of methyl groups (MTHFR and MTRR) and of the enzymes that CinicalTrials.gov, numbers NCT00502047 and NCT00511420.
catalyze the transfer of these methyl groups to DNA (DNMTs)
affect the relationship between cocoa consumption and DNA
methylation. Furthermore, we also evaluated in vitro the effect of a Table 1. Nutritional composition of cocoa cream (13 g
cocoa polyphenol extract treatment on the mRNA expression dose)*.
levels of these five genes related to the DNA methylation pathway
(DNMT1, DNMT3a, DNMT3b, MTHFR and MTRR) using
peripheral blood mononuclear cells (PBMCs) from humans. Content per dose
Energy, Kcal 77
Materials and Methods
Carbohydrate, g 6.6
The protocols for this trial and supporting CONSORT checklist Protein, g 0.2
are available as supporting information; see Checklist S1, Protocol
Total fat, g 6.0
S1 and Protocol S2.
Saturated fat, g 1.2
Stearic fatty acid, g 0.3
Subjects
The population sample for the present study was selected from Monounsaturated fat, g 3.0
earlier studies [21,22]. The population sample included: a) control Polyunsaturated fat, g 1.4
group of 104 individuals from 254 participants [21]; b) cocoa n-6, g 3.0
product group of 110 participants [22]. The participants were Fiber, g 0.3
recruited between April 2005 and June 2007. Inclusion criteria
Vitamin E, mg 2.0
were community-dwelling men and women .20 years of age, with
pre-hypertension (systolic BP: 120–139 mmHg or diastolic BP: SFA/USFA 1:3.5
80–89 mmHG); stage 1 hypertension (systolic BP: 140– MUFA/PUFA 1:2.06
159 mmHg or diastolic BP: 90–99 mmHg); LDL-C SFA/MUFA 1:1.14
$3.35 mmol/L [130 mg/dL] and #4.88 mmol/L [189 mg/ SFA/PUFA 1:2.35
dL]; and at least one other major CVD risk factor such as age
.45 years in men and .55 in women, smoking habit, low HDL-C Kcal: kilocalories; g: gram; n-6: omega 6 (linoleic acid); SFA: saturated fatty acids;
#1.03 mmol/L [40 mg/dL] in men and #1.19 mmol/L [46 mg/ USFA: unsaturated fatty acids; MUFA: monounsaturated fatty acids; PUFA:
polyunsaturated fatty acids.
dL] in women, or family history of premature heart disease. *Nutrient composition calculated from data provided by the manufacturers and
Exclusion criteria included diabetes mellitus, any chronic disease, with USDA food composition tables.
hypolipemic treatment, TG .3.97 mmol/L [350 mg/dL], BMI doi:10.1371/journal.pone.0065744.t001
Global DNA methylation detection the 5mC content between duplicate samples was greater than
Genomic DNA extraction and digestion. The global 0.3%.
methylation levels of peripheral leukocyte DNA from blood The area of each peak was used for the quantitative analysis.
samples and cultured PBMCs were evaluated by measuring the The sample concentrations were calculated from the curves using
5mC content using reversed-phase HPLC. Blood samples were a linear regression, and the level of 5mC in the DNA samples was
obtained from the treated and control subjects at week 2, and then expressed as a percentage of the level of dC, which was
genomic DNA was isolated from the peripheral leukocytes using a calculated using the following equation: %5mC = [5mC/
cell package commercial kit (Servicios Hospitalarios, Spain). DNA (dC+5mC)]6100.
was extracted from the PBMCs using a DNeasyH Blood & Tissue
kit (Qiagen, S.G. Servicios Hospitalarios S.L., Barcelona, Spain). Polymorphisms analysis
The DNA was quantified using a NanoDrop 1000 Spectropho- The analyzed SNPs were selected using two criteria. Because
tometer (ThermoFisher Scientific, Wilmington DE, USA). there are many SNPs in each of the three DNMTs, the majority of
DNA digestion was performed essentially as described elsewhere which are located in introns with unknown functional effects, we
[23,24] but with a few modifications. Briefly, 30 ml of genomic used HapMap (http://hapmap.ncbi.nlm.nih.gov/) to select hap-
DNA was treated with RNase A (Sigma-Aldrich, St. Louis, MO, lotype tagging SNPs to include in the analysis (Tagger-Pairwise,
USA) and RNase T1 (Sigma) at final concentrations of 100 U/mL Hap Map Data Phase III/Rel#2, Feb09, on NCBI B36 assembly,
and 2000 U/mL, respectively, and incubated at 37uC for 2 h. The dbSNP b126 for the CEU population, r2 threshold $0.8 and
DNA was then extracted with phenol-chloroform, precipitated by MAF$0.10). In addition, to select polymorphisms in genes that
sodium acetate (Sigma) and absolute ethanol, and washed twice encode the MTHFR and MTRR enzymes, we used PubMed to
with 70 per cent ethanol. Afterwards, the DNA was resuspended identify SNPs that had previously been reported in the literature to
with 30 ml of 16 deoxyribonuclease I digestion buffer and 10 U of have functional effects. To ensure a wide number of subjects who
DNaseI (New England Biolabs, IZASA S.A., Barcelona, Spain), are homozygous for the rare alleles, we selected only those
and the mixture was incubated overnight at 37uC. Sodium acetate polymorphisms with a minor allele frequency in Europeans of at
(pH 5.2), zinc sulphate, and nuclease P1 (Sigma) were then added, least 10%, as indicated by the SNP public database (dbSNP;
and the mixture was incubated for an additional 7 h at 37uC. After http://www.ncbi.nlm.nih.gov/sites/entrez?db = snp). Finally, a
the digestion, 1/10 volume of 106 alkaline phosphatase buffer total of two MTHFR (by rs number: rs180133, rs180131), two
(10 mM Tris-HCl, pH 8.0, 50 mM KCl, 1 mM MgCl2, 0.1 mM MTRR (rs1801394, rs1532268), three DNMT1 (rs2162560,
ZnCl2) and 10.8 U of alkaline phosphatase were added to the rs759920, rs7253062), four DNMT3A (rs2304429, rs2289195,
DNA solution. Additionally, 2 ml of tetrahydrouridine (Calbio- rs13002567, rs734693) and three DNMT3B (rs998382,
chem, EMD Biosciences, La Jolla, CA, USA) was added to prevent rs4911263, rs2424932) SNPs were included.
the adventitious formation of dU by cytidine deaminase [25], and
Genotyping of the selected polymorphisms was performed using
the solution was incubated overnight at 37uC. Finally, DNA
the Iplex Gold Sequenom technology (coordinacion.cegen@upf.
hydrolysate was filtered using a 0.45-mm filter (Millipore,
edu).
Carrigtwohill, Co. Cork, Ireland), and the samples were stored
at 280uC until analysis.
Generation of Standard curves and HPLC In vitro studies with PBMCs
analysis. Commercially available (TCI Europe, Borenveld- Cocoa extract. The cocoa extract used for this study was
gewes, Belgium) nucleotide standards for 29-deoxyadenosine (dA; produced from cocoa nibs. The total polyphenol content of this
D0046), 29-deoxyguanosine (dG; D0052), thymidine (dT; T0233), cocoa extract, measured using Folin-Ciocalteu’s method, was
and 29-deoxycytidine (dC; D3583) at 100 mM and 29-deoxyuridine 302.5 mg/g. The majority of the polyphenols were monomeric
(dU) and 29-deoxy-5-methylcytidine (5mC; D0046) at 5 mM were (catechin, 2.374 mg/g and epicatechin, 5.638 mg/g), dimeric
used to prepare a mixed sample of all six nucleotides to generate (13.20 mg/g), trimeric (6.216 mg/g), tetrameric (1.571 mg/g) or
calibration curves for each sample batch. Linear calibration curves oligomeric (5–9 units, 0.3 mg/g) proanthocyanidins (measured
were obtained in the concentration range of 2.0 to 100.0 mM for using ultraperformance liquid chromatography MS). The cocoa
dC, dG, dT and dA and in the range of 0.1–5.0 mM for 5mC and extract used in the in vitro experiments had similar characteristics
dU, spanning the expected concentration ranges of the samples. to that used in the human intervention study.
The daily precision was examined by analysing the deoxynucleo- PBMCs preparation and treatment. PBMCs were isolated
side standards (100 mM dC, dG, dT, and dA and 5 mM 5mC and from 20 mL EDTA anticoagulated blood of six volunteers by
dU) at the start and end of every experiment. density gradient centrifugation using Ficoll-PaqueTM PLUS (GE
The 5mC content was measured using an Agilent 1100 Series Healthcare Bio-Sciences AB, Uppsala, Sweden) and stored in
liquid chromatograph (Agilent Technologies) equipped with a liquid nitrogen. The cells were thawed by slow reconstitution with
diode array detector (Agilent Technologies) at room temperature. RPMI 1640 medium supplemented with 10% fetal bovine serum,
The column was an Eclipse Plus C18 column (25064.6 mm, 2% L-glutamine and 1% penicillin-streptomycin (Lonza, Barce-
5 mm) with a ZORBAX Eclipse Plus-C18 guard column lona, Spain). The PBMCs fraction was adjusted to 1.06106 cells/
(12.564.6 mm, 5 mm) (Agilent Technologies). The mobile phase mL, and 1 mL of cells was aliquoted into each well of the 12-well
was 50 mM ammonium orthophosphate, pH 4.1, which was flat-bottom plates.
prepared by dissolving 50 mmol of diammonium orthophosphate To evaluate the effect of the cocoa extract on mRNA
in 50 mM orthophosphoric acid with a subsequent adjustment of expression, the cells were treated with 10 ml of a cocoa extract
the pH to 4.14. The flow rate was 1 mL/min, with a 20 ml with two final concentrations (25 mg/L and 50 mg/L) dissolved in
injection volume, and the sample detection was set to 275 nm 5% DMSO. The cells treated only with vehicle (5% DMSO)
(UV). served as controls. The cells were incubated at 37uC in 5% CO2
For each sample, at least two separate aliquots of DNA were for 3 h and 24 h. The cell culture experiments were performed in
digested and chromatographed. All DNA samples were analyzed triplicate and repeated two times. After treatment, the cells were
in duplicate, and we eliminated samples for which the difference in obtained by centrifugation at 7006 g for 10 min at 4uC.
To evaluate the effect of the cocoa extract on the global DNA genotypes, ANCOVA was used for all of the analyses with the
methylation, the PBMCs were treated with the highest concen- %5mC defined as the dependent variable and the treatment and
tration of cocoa extract (50 mg/L) for 72 h. The detection of the covariates included in the model. The analyses were performed
global DNA methylation in the PBMCs cultures was performed as under the recessive, dominant, and codominant models. For the
described above. analyses of the mRNA expression levels and the global DNA
Isolation of total RNA and quantitative real-time methylation levels in PBMCs, a one-way ANOVA followed by a
PCR. The total RNA from the PBMCs samples was extracted Tukey test or a two-way ANOVA was used.
using a combination of Tripure Reagent (Roche Diagnostics, With 104 and 110 subjects per group, the study has 90% power
Barcelona, Spain) and an RNeasyH MiniEluteTM Cleanup Kit to detect a difference of at least 0.5% in %5mC assuming a SD of
(Qiagen, S.G. Servicios Hospitalarios S.L., Barcelona, Spain), 0.800, with a global two-sided alpha of 5%.
dissolved in 22 ml of RNase free water and stored at 280uC. The The experimental results with continuous variables are
RNA yield was quantified using a NanoDrop 1000 Spectropho- expressed as the mean 6 SEM or the mean 6 SD, whereas the
tometer (ThermoFisher Scientific, Wilmington DE, USA). human data results of %5mC are expressed as the least-square
The reverse transcription reactions were performed with 400 ng mean 6 SEM or the least-square mean and 95% confidence
of RNA that was reverse transcribed to cDNA using the High- intervals (95% CI) for each genotype.
Capacity cDNA Reverse Transcription Kit (Applied Biosystems) All statistical analyses were performed with SPSS version 17.0
following the manufacturer’s protocol. An analysis of the relative software (SPSS Inc., Chicago, IL, USA). Statistical significance
gene expression for DNMT1, DNMT3A, DNMT3B, MTHFR and was considered at p#0.05. For genotypes analysis, we applied
MTRR was performed using real-time quantitative PCR in a 7300 step-down Sidak adjustment to correct for multiple testing and the
Real-Time PCR System (Applied Biosystems). For the amplifica- two p values are presented.
tion reactions, the TaqMan Gene Expression Master Mix and
Inventoried Gene Assay Products (Applied Biosystems) containing Results
2 gene-specific primers and 1 TaqMan MGB probe (6-FAM dye-
labelled) from Applied Biosystems were used (assay IDs DNMT1: Global DNA methylation in human peripheral leukocyte
Hs00154749_m1, DNMT3A: Hs01027166_m1, DNMT3B: DNA
Hs00171876_m1, MTHFR: Hs00195560_m1, MTRR: We measured the global peripheral leukocyte DNA methylation
Hs00985015_m1). The thermal cycle conditions were as follows: content in a population that consumed 6 g/d of cocoa in the form
first cycle at 50uC for 2 minutes followed by 95uC for 10 minutes, of chocolate for two weeks (treated subjects) and in a population
and 40 cycles at 95uC for 15 seconds followed by 60uC for that did not consume any chocolate and thus served as controls.
1 minute. A no-template control was included in each assay. Table 2 shows the characteristics of the participants of the study.
RPLP0 (Hs99999902_m1) was used as an endogenous control, and No significant differences in age, gender, weight or body mass
a vehicle control was used as a calibrator. The expression levels of index were observed between the two groups.
the target transcripts in each sample were calculated using the
We observed a statistically significant difference in the
comparative Ct method (22DCt formula) after normalization to
methylation status between the two groups analyzed. As shown
RPLP0 mRNA.
in Figure 1, the %5mC of the peripheral leukocyte DNA was
Assay of cell viability. The cocoa extract toxicity on
significantly lower in the treated subjects than in the control
cultured PBMCs was determined by measuring the amount of
subjects (2.99160.366 vs. 3.90960.380, p,0.001). In our study,
intracellular lactate dehydrogenase (LDH) in the culture superna-
there were no statistically significant differences between men and
tants. The PBMCs were incubated in 12-well plates at a density of
woman in the treated or the control group, and no correlation was
16106 cells/well with different concentrations of cocoa extract (1,
observed between the methylation value and the subject’s age
10, 50, 100, and 250 mg/L) for 24 h. After incubation, the LDH
(data not shown).
activity in the medium was measured by analyzing the NAD
Moreover, in the treated group, the %5mC levels were higher in
reduction using an LDH kit (QCA, Barcelona, Spain) according to
the homozygote G subjects (mean (95% CI); 3.117 (2.991 to
the manufacturer’s instructions and measuring the absorbance at
3.242)) than in the carriers of the allele A (2.945 (2.869 to 3.021))
340 nm.
(p = 0.022) of the rs1801394 polymorphism in the MTRR gene,
The LDH activity was expressed as the percentage of the total
although the corrected p value was not statistically significant
LDH cellular released following the addition of (0.1%) Triton X-
(p = 0.268). No association was observed between the others
100. The survival of the vehicle-treated control cultures receiving
polymorphisms analyzed and the %5mC levels in total treated
no cocoa extracts was defined as 100%, and that of the treated
groups was expressed as a percentage of the control group. The
results are given as the mean (6 SD) of different incubations, each Table 2. Characteristics of participants.
performed in triplicate.
Figure 2. Relative gene expression levels of DNA methyltransferases in in vitro PBMC. Relative mRNA levels of DNA methyltransferases
DNMT1 (a), DNMT3a (b), and DNMT3b (c) after 24 h treatment with cocoa extract (25 mg/L or 50 mg/L) or with vehicle. Results represent mean6SEM
expression levels normalized to RPLP0. DNMT1, DNMT3a and DNMT3b expression were significantly reduced by cocoa treatment relative to control
treated with vehicle alone (one-way ANOVA and Tukey’s post hoc comparison). abMean values not sharing the same letters were significantly
different. *p#0.05; ##0.001.
doi:10.1371/journal.pone.0065744.g002
DNA methylation compared with those without the MTRR or epigenetic process. Therefore, we performed in vitro experiments
MTHFR polymorphisms. MTHFR and MTRR play central roles by treating PBMCs from subjects with a cocoa extract with its
in the biotransformation of folate to form S-adenosylmethionine, naturally occurring combination of polyphenols. In our study, low
the universal methyl donor in cells [3,4]. Additionally, several concentrations of polyphenols were sufficient to obtain a
studies have shown that these two polymorphisms are associated significant reduction in the mRNA levels of the DNMTs, MTHFR
with reduced enzyme activity [29,30], which may influence the and MTRR genes. Interestingly, the reduction in gene expression
pool of methyl-donor molecules and, therefore, DNA methylation was observed with each of the three main DNMTs analyzed,
[31,32]. Our results suggest that cocoa consumption produces a which is consistent with other reports that have shown that several
more potent effect on the reduction of DNA methylation levels in dietary polyphenols, such as caffeic acid, chlorogenic acid, EGCG,
humans with normal activities of MTHFR and MTRR enzymes. catechin, epicatechin and genistein, are inhibitors of all three
We also observed an association between cocoa consumption and DNMTs [10,13,33]. Because the three DNMTs have been shown
the DNMT3B A38946G polymorphism on global peripheral to exhibit some level of both maintenance and de novo methylation
leukocyte DNA methylation in men. This polymorphism is located in vitro [34], our results in PBMCs suggest that the hypomethyla-
in an intronic region of the DNMT3B gene and has never been tion effect of cocoa could be produced, in part, through the down-
studied previously. The results of our study suggest that this regulation of these genes.
polymorphism may have a functional effect. In our study, the ability of cocoa extract to decrease mRNA
In the second part of the study, we tested the hypothesis that expression was correlated with its ability to decrease global DNA
cocoa could decrease the global peripheral leukocyte DNA methylation in PBMCs, although the differences were not
methylation via the regulation of key genes involved in this statistically significant. These findings reinforce our hypothesis
Figure 3. Relative gene expression levels of MTHFR and MTRR in in vitro PBMCs. Relative mRNA levels of MTHFR (a) and MTRR (b) after 24 h
treatment with cocoa extract (25 mg/L or 50 mg/L) or with vehicle alone. Results represent mean6SEM expression levels normalized to RPLP0.
MTHFR and MTRR expression were significantly reduced by cocoa treatment relative to control treated with vehicle alone (one-way ANOVA and
Tukey’s post hoc comparison). abcMean values not sharing the same letters were significantly different. *p#0.05; #p#0.001.
doi:10.1371/journal.pone.0065744.g003
References
1. Link A, Balaguer F, Goel A (2010) Cancer chemoprevention by dietary 7. Zhao J, Goldberg J, Bremmer JD, Vaccarino V (2012) Global DNA methylation
polyphenols: promising role for epigenetics. Biochem Pharmacol 80: 1771–1792. is associated with insulin resistance: a monozygotic twin study. Diabetes 61: 542–
2. Li Y, Tollefsbol TO (2010) Impact on DNA methylation in cancer prevention 546.
and therapy by bioactive dietary components. Curr Med Chem 17: 2141–2151. 8. Pearce MS, McConnell JC, Potter C, Barrett LM, Parker L, et al. (2012) Global
3. Fang JY, Xiao SD (2003) Folic acid, polymorphisms of methyl-group LINE-1 DNA methylation is associated with blood glycaemic and lipid profiles.
metabolism genes, and DNA methylation in relation to GI carcinogenesis. Int J Epidemiol 41: 210–217.
J Gastroenterol 38: 821–829. 9. Duthie SJ (2011) Epigenetic modifications and human pathologies: cancer and
4. Das PM, Singal R (2004) DNA methylation and cancer. J Clin Oncol 22: 4632– CVD. Proc Nutr Soc 70: 47–57.
4642. 10. Lee WJ, Shim JY, Zhu BT (2005) Mechanisms for the inhibition of DNA
5. Kim M, Long TI, Arakawa K, Wang R, Yu MC, et al. (2010) DNA methylation methyltransferases by tea catechins and bioflavonoids. Mol Pharmacol 68: 1018–
as a biomarker for cardiovascular disease risk. PLoS One 5: 1–8. 1030.
6. Stenvinkel P, Karimi M, Johansson S, Axelsson J, Suliman M, et al. (2007) 11. Fang MZ, Chen D, Sun Y, Jin Z, Christman JK, et al. (2005) Reversal of
Impact of inflammation on epigenetic DNA methylation- a novel risk factor for hypermethylation and reactivation of p16INK4a, RARb, and MGMT genes by
cardiovascular disease?. J Int Med 261: 488–499. genistein and other isoflavones from soy. Clin Cancer Res 11: 7033–7041.
12. Fang M, Chen D, Yang CS (2007) Dietary polyphenols may affect DNA 27. Wang LS, Arnold M, Huang YW, Sardo C, Seguin C, et al. (2011) Modulation
methylation. J Nutr (suppl 1): S223–S228. of genetic and epigenetic biomarkers of colorectal cancer in humans by black
13. Lee WJ, Zhu BT (2006) Inhibition of DNA methylation by caffeic acid and raspberries: a phase I pilot study. Clin Can Res 17: 598–610.
chlorogenic acid, two common catechol-containing coffee polyphenols. Carci- 28. Cho NY, Kim BH, Choi M, Yoo EJ, Moon KC, et al. (2007) Hypermethylation
nogenesis 27: 269–277. of CpG island loci and hypomethylation of LINE-1 and Alu repeats in prostate
14. Liu Z, Xie Z, Jones W, Pavlovicz RE, Liu S, et al. (2009) Curcumin is a potent adenocarcinoma and their relationship to clinicopathological features. J Pathol
DNA hypomethylation agent. Bioorg Med Chem Lett 19: 706–709. 211: 269–277.
15. Day JK, Bauer AM, desBordes C, Zhuang Y, Kim BE, et al. (2002) Genistein 29. Weisberg IS, Jacques PF, Selhub J, Bostom AG, Chen Z, et al. (2001) The
alters methylation patterns in mice. J Nutr 132(suppl8): 2419S–2423S. 1298ARC polymorphism in methylenetetrahydrofolate reductase (MTHFR): in
16. Dolinoy DC, Weidman JR, Waterland RA, Jirtle RL (2006) Maternal genistein vitro expression and association with homocysteine. Atherosclerosis 156: 409–
alters coat color and protects Avy mouse offspring from obesity by modifying the 415.
fetal epigenome. Environ Health Perspect 114: 567–572. 30. Olteanu H, Munson T, Banerjee R (2002) Differences in the efficiency of
17. Guerrero-Bosagna CM, Sabat P, Valdovinos FS, Valladares LE, Clark SJ (2008) reductive activation of methionine synthase and exogenous electron acceptors
Epigenetic and phenotypic changes result from a continuous pre and post natal between the common polymorphic variants of human methionine synthase
dietary exposure to phytoestrogens in a experimental population of mice. BMC reductase. Biochemistry 41: 13378–13385.
Physiol 8: 17–35. 31. deVogel S, Wouters KA, Gottschalk RW, van Schooten FJ, de Goeij AF, et al.
18. Morey-Kinney SR, Zhang W, Pascual M, Greally JM, Gillard BM, et al. (2009) (2009) Genetic variants of methyl metabolizing enzymes and epigenetic
Lack of evidence for green tea polyphenols as DNA methylation inhibitors in regulators: associations with promoter CpG island hypermethylation in
murine prostate. Cancer Prev Res 2: 1065–1075. colorectal cancer. Cancer Epidemiol Biomarkers Prev 18: 3086–3096.
19. Pérez-Jiménez J, Neveu V, Vos F, Scalbert A (2010) Identification of the 100 32. Friso S, Girelli D, Trabetti E, Olivieri O, Guarini P, et al. (2005) The MTHFR
richest dietary sources of polyphenols: an application of the Phenol-Explorer 1298A.C polymorphism and genomic DNA methylation in human lympho-
database. Eur J Clin Nutr 64 (suppl3): S112–S120. cytes. Cancer Epidemiol Biomarkers Prev 14: 938–943.
20. Rimbach G, Melchin M, Moehring J, Wagner AE (2009) Polyphenols from 33. Majid S, Dar AA, Shahryari V, Hirata H, Ahmad A, et al. (2010) Genistein
Cocoa and vascular health-a critical review. In J Mol Sci 10: 4290–4309. reverses hypermethylation and induces active histone modifications in tumor
21. Solà R, Bruckert E, Valls RM, Narejos S, Luque X, et al. (2010) Soluble fibre suppressor gene B-cell translocation gene 3 in prostate cancer. Cancer 116: 66–
(Plantago ovate husk) reduces plasma low-density lipoprotein (LDL) cholesterol, 76.
triglycerides, insulin, oxidized LDL and systolic blood pressure in hypercholes- 34. Pradhan S, Bacolla A, Wells R, Roberts RJ (1999) Recombinant human DNA
terolaemic patients: A randomized trial. Atherosclerosis 211: 630–637. (cytosine-5) methyltransferase. I. Expression, purification, and comparison of de
22. Solà R, Valls RM, Godàs G, Perez-Busquets G, Ribalta J, et al. (2012) Cocoa, novo and maintenance methylation. J Biol Chem 274: 33002–33010.
hazelnuts sterols and soluble fiber cream reduces lipids and inflammation 35. Kato K, Long NK, Makita H, Toida M, Yamashita T, et al. (2008) Effect of
biomarkers in hypertensive patients: a randomized controlled trial. PLoS One 7: green tea polyphenol on methylation status of RECK gene and cancer cell
e31103. invasion in oral squamous cell carcinoma cells. Br J Cancer 99: 647–654.
23. Chen KL, Wang SSS, Yang YY, Yuan RY, Chen RM (2009) The epigenetic 36. Pandey M, Shukla S, Gupta S (2010) Promoter demethylation and chromatin
effects of amyloid-b1–40 on global DNA and neprilysin genes in murine cerebral remodelling by green tea polyphenols leads to re-expression of GSTP1 in human
endothelial cells. Biochem Biophys Res Commun 378: 57–61. prostate cancer cells. Int J Cancer 126: 2520–2533.
24. Ramsahoye BH (2002) Measurement of genomic wide DNA methylation by 37. Fini L, Selgrad M, Fogliano V, Graziani G, Romano M, et al. (2007) Annurca
reversed-phase high-performance liquid chromatography. Methods 27: 156– apple polyphenols have potent demethylating activity and can reactivate silenced
161. tumor suppressor genes in colorectal cancer cells. J Nutr 137: 2622–2628.
25. Dong M, Dedon PC (2006) Relatively small increases in the steady-state levels of 38. Buitrago-Lopez A, Sanderson J, Johnson L, Warnakula S, Wood A, et al. (2011)
nucleobase deamination products in DNA from TK6 cells exposed to toxic levels Chocolate consumption and cardiometabolic disorders: systematic review and
of nitric oxide. Chem Res Toxicol 19: 50–57. meta-analysis. BMJ 343: d4488.
26. Hammoud SS, Cairns BR, Carrell DT (2013) Analysis of gene-specific and 39. Ried K, Sullivan T, Fakler P, Frank OR, Stocks NP (2010) Does chocolate
genome-wide sperm DNA methylation. Methods Mol Biol 927: 451–458. reduce blood pressure? A meta-analysis. BMC Med 8: 39–49.